Doubled haploid plant production from unpollinated ovules of sugar beet (Beta vulgaris L.)

Transcription

1 Plant Cell Reports (2000) 19: Q Springer-Verlag 2000 CELL BIOLOGY AND MORPHOGENESIS S. Gürel 7 E. Gürel 7 Z. Kaya Doubled haploid plant production from unpollinated ovules of sugar beet (Beta vulgaris L.) Received: 25 October 1999 / Revision received: 18 May 2000 / Accepted: 29 May 2000 Abstract The effects of cold pretreatment, AgNO 3 and activated charcoal on haploid plant production from unpollinated sugar beet ovules were investigated. Both cold pretreatment and the addition of charcoal increased the frequency of embryo formation, whereas AgNO 3 decreased or completely inhibited it. Colchicine (50, 100, 150 or 500 mg l 1 ) and trifluralin (1.7, 3.4 or 5.0 mg l 1 ) for 12, 24, 36 or 48 h were compared in agar-solidified, agarose-solidified or liquid media. Although colchicine gave a higher doubling rate (25.3%) than trifluralin (18.2%), the difference was not significant. Both agents were more effective when used in liquid (29.1%) than agarose-solidified medium (20.7%) and agar-solidified medium (15.4%). A treatment duration of 48 h was significantly more effective (27.5%) than 12 h (13.6%) but it was not different from 24 h (16.3%) or 36 h (18.6%). Keywords Beta vulgaris L. 7 Ovule culture 7 Haploid plant production 7 Chromosome doubling Communicated by S.A. Merkle S. Gürel Sugar Institute, Department of Plant Breeding, Etimesgut, Ankara, Turkey E. Gürel Abant Izzet Baysal University, Department of Biology, Bolu, Turkey Z. Kaya (Y) Middle East Technical University, Department of Biology, Ankara, Turkey kayaz6metu.edu.tr Fax: c Present address: S. Gürel Middle East Technical University, Department of Biology, Ankara, Turkey Abbreviations APM: Amiprophosmethyl 7 BAP: 6-Benzylaminopurine 7 MS: Murashige and Skoog (1962) 7 NAA: a-naphthaleneacetic acid Introduction Doubled haploid methodologies permit the production of a series of homozygous breeding lines from a given genotype. In sugar beet (Beta vulgaris L.), many different approaches have been suggested for the induction of haploid and dihaploid plants. Classical techniques such as natural polyembryony and crosses between diploid and tetraploid lines or with wild species yielded very low numbers of haploids (Van Geyt et al. 1987). In vitro androgenesis has been attempted numerous times (Van Geyt et al. 1985) but with very limited success. Gynogenesis has been the most successful method for the production of haploid plants (Bossoutrot and Hosemans 1985; D Halluin and Keimer 1986; Lux et al. 1990; Hansen et al. 1994, 1995, 1998). Colchicine is known to inhibit mitosis in a wide variety of plant and animal cells by interfering with the structure of the mitotic spindle, but it is highly toxic and often induces chimeric plants (Rao and Suprassana 1996). Other antimitotic agents such as trifluralin, oryzalin, pronamide and amiprophosmethyl (APM) have also been tested in sugar beet (Hansen et al. 1998) and other species including maize (Wan et al. 1991), rape (Zhao and Simmonds 1995) and wheat (Hassawi and Liang 1991). The present study examined the effects of cold pretreatment of unopened flower buds and the addition of charcoal or silver nitrate (AgNO 3 ) to the culture medium on the production of haploid plants from cultured ovules. Also, colchicine was compared with an alternative antimitotic agent (trifluralin) for chromosome doubling, using three different culture techniques (solid-agar, solid-agarose and liquid medium).

2 1156 Materials and methods Haploid plant production Ovule explants of 13 diploid male fertile sugar beet breeding lines (listed in Tables 1 and 2) were used for haploid plant production. Plants were grown in the field during the normal season in Ankara (Turkey) and unopened flower buds were excised from bolted plants in June and July, and surface sterilised in 2% NaOClc0.03% Tween 20 for 20 min followed by several rinses with sterile distilled water. Ovules were isolated by using a dissecting microscope under sterile conditions and cultured on full-strength MS (Murashige and Skoog 1962) basal salt medium supplemented with 10% sucrose and 0.75% agar (Oxoid No. 3). For haploid embryo induction, two sets of experiments were carried out. In the first set, isolated ovules were cultured on MS medium containing 1.0 or 2.0 mg l 1 BAP alone or in combination with a cold pretreatment of flower buds at 4 7C for 5 days in the dark (Table 1). In the second set, ovules were cultured on MS medium containing 2.0 mg l 1 BAP alone or in combination with 2.5 or 5.0 mg l 1 AgNO 3, or with 0.5% activated charcoal (Merck, Germany; Table 2). Embryos developed from the cultured ovules were then transferred to MS medium containing 0.5 mg l 1 BAP and 3% sucrose for further growth. Cultures were incubated at 27B2 7C with a 16-h photoperiod under warm white fluorescent tubes (40 50 mmol m 2 s 1 ). Experiments were repeated three times, and every treatment of each replication used 30 ovules. The number of ovule explants giving rise to haploid plants was recorded and expressed as the mean percentage of haploid plants per treatment, excluding the spontaneously doubled plants. Treatment means for percentage of haploid plants and for percentage of plants showing various ploidy levels were estimated using the PROC MEANS procedure of the SAS statistical package (SAS replace with: Institute 1988). Analyses of variance treatment effects on these variables were carried out using the PROC ANOVA procedure with the Duncan s multiple range test option of the SAS statistical package (SAS Institute 1988). Chromosome doubling Haploid plants obtained from lines M23 and M64 only were used for chromosome doubling experiments. After excising the roots, shoots taller than 1 cm were either (1) incubated on medium solidified with 0.7% agar, (2) embedded in medium solidified with 0.3% agarose, or (3) immersed in liquid medium, all containing colchicine (50, 100, 150 or 500 mg l 1 ) or trifluralin (1.7, 3.4 or 5.0 mg l 1 ) for 12, 24, 36 or 48 h at 27B2 7C (Table 3). All media were composed of full-strength MS and supplemented with 1.0 mg l 1 BAP and 3% sucrose. Treated shoots were then rinsed with sterile distilled water and transferred to an agar-solidified MS medium supplemented with 1.0 mg l 1 BAP. Following the ploidy determination, doubled haploid plants were transferred to MS medium containing 3.0 mg l 1 NAA for rooting and finally transferred to soil. Experiments were repeated three times, and every treatment of each replication used 15 haploid plants. Numbers of diploid (doubled, 2n), tetraploid (4n) and mixoploid (nc2n and/or 2nc4n) plants were recorded and expressed as the mean % of plants per treatment (Tables 3, 4). Treatment means for percentage of haploid plants and for percentage of plants showing various ploidy levels were estimated using the PROC MEANS procedure of the SAS statistical package (SAS Institute 1988). Analyses of variance treatment effects on these variables were carried out using the PROC ANOVA procedure with the Duncan s multiple range test option of the SAS statistical package (SAS Institute 1988). Table 1 The effects of 6-benzylaminopurine (BAP, mg l 1 ) in combination with or without cold pretreatment on haploid plant production (expressed as the mean percentage of haploid plants per treatment). Means with the same letter are not significantly different at P~0.05 Lines BAP (1.0) BAP (2.0) Cold (4 7C) cbap (1.0) Cold (4 7C) Means M a 6.7 a 13.3 a 13.3 a 10.1 a M b 0.0 b 10.3 a 11.0 b 5.3 b M a 5.7 a 6.0 b 4.3 d 5.1 b M22 a a 2.0 c 3.3 d 2.7 c M a 4.3 (10.0) b a 7.0 b 8.0 c 5.7 b M b 0.0 b 7.2 b 8.3 c 3.9 c M b 0.0 b 4.3 c 9.0 c 3.3 c Means 2.2 b 2.8 b 7.2 a 9.6 a a No data due to infection b The figure in parentheses is the mean percentage of plants which doubled spontaneously Table 2 The effects of BAP (mg l 1 ) alone or in combination with AgNO 3 (mg l 1 ) or charcoal (%) on haploid plant production (expressed as the mean percentage of haploid plants per treatment). Means with the same letter are not significantly different at P~0.05 Lines BAP (2.0) AgNO 3 (2.5) AgNO 3 (5.0) Charcoal (0.5) Means a M c 1.7 c 0.0 c 16.7 b 4.6 a M b 1.7 c 1.7 b 18.3 b 6.3 a M a 3.3 b 0.0 c 10.0 a 4.6 a M b 3.3 b 0.0 c 10.0 a 4.2 a M b 5.0 (1.7) b a 1.7 b 13.3 a 5.8 a M a 3.3 b 6.7 a 8.3 a 5.8 a Means 3.3 b 3.1 b 1.7 b 12.8 a a The differences were not significant b The figure in parentheses is the mean percentage of plants which doubled spontaneously

3 Table 3 Effects of different lines, culture media, chromosome doubling agents, their concentrations and treatment durations on doubled haploid plant production (expressed as the mean percentage of doubled haploid plants per treatment) Lines Medium Duration (h) Colchicine (mg l 1 ) Trifluralin (mg l 1 ) M23 Agar Agarose Liquid M64 Agar Agarose Liquid Ploidy determination Young leaves of 45- to 50-day-old shoots were soaked in 8- hydroxyquinoline (0.002 mol l 1 ) for 4 h, followed by several rinses with tap water. Treated leaves were then fixed in a 2 :1 fixative solution (two parts 96% ethanol: one part HCl) for 15 min. The leaves were washed thoroughly with tap water to remove the fixative and were kept in distilled water at room temperature. A small piece of the leaf tissue was excised and placed on a glass slide. After applying a drop of 3% aceto-orcein, the tissue was squashed before counting chromosomes under a light microscope. Results and discussion Production of haploid plants Certain physical treatments applied to ovules, plants or plant parts from which ovule explants have been taken prior to culture may have a strong influence on embryo induction. Cold pretreatment combined with 2.0 mg l 1 BAP produced more embryos (9.6%) than with 1.0 mg l 1 BAP (7.2%), though the difference was not significant (Table 1). However, across all the lines, the cold pretreatment combined with either level of BAP induced significantly more embryos. This result is in contrast to the findings of D Halluin and Keimer (1986) who reported that cold pretreatment of flower buds at 4 7C for 1 week had no effect on haploid embryo induction. Incorporation of activated charcoal in the culture medium stimulates androgenesis in some systems, presumably by removing the growth inhibitors from the medium (Reinert and Bajaj 1977). D Halluin and Keimer (1986) reported that the inclusion of charcoal was more effective in embryo induction when very young sugar beet ovules were used. Our results showed that the addition of 0.5% charcoal to the culture medium increased embryo formation significantly, with a mean of 12.8% compared to % in other treatments (Table 2). The inclusion of AgNO 3 either decreased haploid embryo induction at 2.5 mg l 1 (except lines M20 and M65) or completely inhibited it at 5.0 mg l 1 (Table 2). Silver nitrate is known to be a potential inhibitor of ethylene action in plants and was shown to promote wheat pollen embryogenesis by blocking the inhibitory effect of endogenous ethylene on embryo induction (Ghaemi et al. 1994). A beneficial effect of AgNO 3 was also reported in Brussels sprouts (Ockendon and Clenaghan 1993). We observed that ovule explants became brown less rapidly in the presence of AgNO 3, suggesting that ethylene-promoted senescence may have been slowed down. In fact, the exact role of silver ions in in vitro organogenesis or embryogenesis is not known and the related reports are rather conflicting. Genotypic variation was evident in the first set of experiments (Table 1). The highest embryo yield was obtained from line M18 (10.1%). The line M18 differed significantly from other lines. Also lines M23 (5.7%), M19 (5.3%), and M21 (5.1%) had significantly higher percentages of haploid plants than lines M22 (2.7%), M101 (3.9%) or M1017 (3.3%). Lines M19, M101 and

4 1158 Table 4 Overall comparisons of different parameters for doubled haploid (2n), tetraploid (4n) and mixoploid (nc2n and/or 2nc4n) plant production (expressed as the mean percentage of plants per treatment). Each parameter is independent from others and the means with the same letters are not significantly different at P~0.05 Parameter Treatment Mean % of Plants Parameter Treatment Mean % of Plants 2n 4n mix. 2n 4n mix. Lines M a 1.2 a 2.2 a Agents Colchicine 25.3 a 2.1 a 1.8 a M a 1.3 a 2.9 a Trifluralin 18.2 a 0.1 a 3.7 a Media Agar 15.4 b 0.7 a 2.3 a Concentrations (mg l 1 ) Col (50) 17.5 c 0.0 c 1.1 b Agarose 20.7 b 1.9 a 4.2 b Col (100) 23.1 b 0.0 c 1.8 b Liquid 29.1 a 1.1 a 1.7 a Col (150) 29.7 a 2.1 b 0.6 b Col (500) 30.8 a 6.5 a 3.8 a Durations (h) b 0.4 b 2.0 a b 0.4 b 3.8 a Tri (1.7) 9.0 b 0.0 a 2.2 b b 0.7 b 1.8 a Tri (3.4) 23.9 a 0.0 a 2.8 b a 3.3 a 3.2 a Tri (5.0) 21.6 a 0.3 a 6.1 a M1017 did not produce embryos at all when they did not receive cold pretreatment, while others, although fewer than those treated with cold, produced some. Genotypic variation has been recognised as a serious problem not only with ovule cultures (Doctrinal et al. 1989; Lux et al. 1990; Hansen et al. 1995) but also with other tissue cultures of sugar beet (Mikami et al. 1989; Gürel 1997). However, genotypic variation among lines was not significant in the second set of experiments (Table 2) when the overall means of all treatments were compared. This could be explained by the interaction observed between lines and treatments (data not shown), indicating that the optimal culture treatment might be variable for different lines. Production of doubled haploid plants Three media (solid-agar, solid-agarose and liquid), two antimitotic agents (colchicine and trifluralin) at different concentrations, four treatment times (12, 24, 36 and 48 h) and two breeding lines (M23 and M64) were compared for their efficiency of chromosome doubling of haploid plants (Table 3). Antimicrotubule herbicides including trifluralin are regarded as effective alternatives to colchicine for chromosome doubling in many species (Hassawi and Liang 1991; Wan et al. 1991; Zhao and Simmonds 1995; Zhao et al. 1996; Hansen et al. 1998). When the overall means of all the treatments tested were compared (Table 4), colchicine was found to result in a higher doubling rate (25.3%) than trifluralin (18.2%), but the difference was not statistically significant. However, it should be noted that the concentrations of trifluralin used ( mg l 1 ) were much lower than those of colchicine ( mg l 1 ). In contrast to our findings, Hassawi and Liang (1991) found that colchicine was far more effective (89%) than oryzalin (7.7%) or trifluralin (18.8%) for doubled haploid plant production in wheat. In a recent study, different antimitotic agents (amiprophosmethyl, pronamide, trifluralin and oryzalin) were tested on sugar beet ovules for doubled haploid plant production (Hansen et al. 1998). These authors demonstrated that APM was the best of the four agents, resulting in a mean of 4.7 doubled plants per 100 ovules cultured. However, it is not possible to say whether any of these agents, including APM, would be better than colchicine, since colchicine was not included in their experiments. In addition, the random distribution of ovule explants from five different sources (three types of sugar beet and two types of fodder beet) over the treatments undermines the reliability of their results, since substantial genotypic variation has been previously reported by the same group (Hansen et al. 1995). Higher concentrations of colchicine (150 and 500 mg l 1 ) produced significantly greater numbers of doubled haploid plants (29.7% and 30.8%, respectively) than did the lower concentrations (17.5% and 23.1% at 50 and 100 mg l 1, respectively). For trifluralin, the highest doubling rate was obtained at 3.4 mg l 1 (23.9%). There were no significant differences between the higher concentrations of trifluralin with regard to production of doubled haploids, but the low concentration significantly reduced double haploid plant production (9.0% at 1.7 mg l 1 ; Table 4). On the other hand, the highest chromosome doubling frequency (62.5%) of all the treatments tested was obtained when haploid plants of line M23 were treated with 3.4 mg l 1 trifluralin for 36 h (Table 3). Treatment durations of both agents also seemed to influence the frequency of diploid plants. A treatment duration of 48 h was significantly more effective (27.5%) than other durations (Table 4). Diploid plant percentages ranged from 13.6% for 12 h to 18.6% for 36 h, though these did not vary significantly (Table 4). However, because of the large interaction between duration and agent concentration (data not shown), it was difficult to determine if the duration times other than 48 h also had significant effects on diploid plant yield. Physical properties of the culture medium were found to be significant for the doubling efficiency of haploid sugar beets. Higher rates of chromosome doubling were obtained when the shoots were immersed in

5 1159 liquid medium (29.1%) than when they were embedded in agarose-solidified medium (20.7%), or cultured on agar-solidified medium (15.4%; Table 4). This could be attributed to a better uptake of the agent which is in direct contact with the meristems in liquid. This finding is partially in agreement with that of Bouvier et al. (1994) who observed that the efficiency of agarosesolidified medium was higher than either agar-solidified or liquid medium for chromosome doubling of haploid apple shoots with colchicine or oryzalin. In contrast to some of the previous reports (Hansen et al. 1995), we observed no significant differences between the lines, mean doubling rates being 22.0% and 22.5% in lines M23 and M64, respectively. Lines M23 and M64 produced similar numbers of tetraploids (4n) and mixoploids (nc2n and/or 2nc4n; Table 4). This was the case for different types of media as well. With regard to treatment duration, 48 h induced more tetraploids than other durations. Colchicine induced more tetraploids than trifluralin: 2.1% compared with 0.1%, respectively. While higher levels of colchicine (150 and 500 mg l 1 ) produced tetraploids at high frequencies (2.1% and 6.5%, respectively), lower levels (50 and 100 mg l 1 ) produced none at all. On the other hand, trifluralin had almost no effect on the induction of tetraploids. The frequency of tetraploids in our study was much lower than that observed in the work of Bossoutrot and Hosemans (1985) in which 10 40% of haploid plants exposed to 100 mg l 1 colchicine produced tetraploids. Such high frequencies might be attributed to the longer treatment durations they employed, from 24 to 120 h. We also observed that survival rates of the shoots decreased with increasing colchicine and trifluralin concentrations (data not shown), and the longer treatments caused shoots to wilt. The higher rates of doubled haploid plants achieved in the present work by the use of liquid and agarosesolidified media, which were employed for the first time in sugar beet, are promising for future studies. Also, colchicine was compared with an alternative antimitotic agent (trifluralin) in sugar beet for the first time and it was found that both agents were equally effective for doubling haploid plants obtained from cultured ovules. Trifluralin, therefore, might be preferred to colchicine since it is much cheaper and has been reported to be less toxic (Zhao et al. 1996). Acknowledgements The authors are grateful to the Sugar Institute of the Turkish Sugar Factories Co. for supporting the study, and to Dr M Erdal of the Sugar Institute for providing the breeding lines. References Bossoutrot D, Hosemans D (1985) Gynogenesis inbeta vulgaris L.: fromin vitro culture of unpollinated ovules to the production of doubled haploid plants in soil. Plant Cell Rep 4: Bouvier L, Fillon FR, Lespinase Y (1994) Oryzalin as an efficient agent for chromosome doubling of haploid apple shootsin vitro. Plant Breed 113: D Halluin K, Keimer B (1986) Production of haploid sugar beets (Beta vulgaris L.) by ovule culture. In: Horn W, Jensen CJ, Odenbach W, Schieder O (eds) Genetic manipulation in plant breeding. de Gruyter, Berlin, pp Doctrinal M, Sangwan RS, Sangwan-Norreel BS (1989)In vitro gynogenesis inbeta vulgaris L.: effect of plant growth regulators, temperature, lines and season. Plant Cell Tissue Organ Cult 17:1 12 Ghaemi M, Sarrafi A, Alibert G (1994) The effects of silver nitrate, colchicine, cupric sulfate and genotype on the production of embryoids from anthers of tetraploid wheat (Triticum turgidum). Plant Cell Tissue Organ Cult 36: Gürel E (1997) Callus and root development from leaf explants of sugar beet (Beta vulgaris L.): variability between cultivars, plants and organs. Turk J Bot 21: Hansen AL, Plever C, Pedersen HC, Keimer B, Andersen SB (1994) Efficientin vitro chromosome doubling duringbeta vulgaris ovule culture. Plant Breed 112:89 95 Hansen AL, Gertz A, Joersbo M, Andersen SB (1995) Shortduration colchicine treatment forin vitro chromosome doubling during ovule culture ofbeta vulgaris L. Plant Breed 114: Hansen AL, Gertz A, Joersbo M, Andersen SB (1998) Antimicrotubule herbicides forin vitro chromosome doubling in Beta vulgaris L. ovule culture. Euphytica 101 : Hassawi DS, Liang GH (1991) Antimitotic agents: effects on double haploid production in wheat. Crop Sci 31 : Lux H, Herrman L, Wetzel C (1990) Production of haploid sugar beet (Beta vulgaris L.) by culturing unpollinated ovules. Plant Breed 104: Mikami T, Sudoh R Nagao, Kinoshita T (1989) Genotypic variation in thein vitro morphogenesis from leaf explants ofbeta vulgaris L. andbeta maritima L. Euphytica 40: Murashige T, Skoog F (1962) A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol Plant 15: Ockendon DJ, Clenagham R (1993) Effect of silver nitrate and 2,4-D on anther culture of Brussels sprouts. Plant Cell Tissue Organ Cult 32:41 46 Rao PS, Suprasanna P (1996) Methods to double haploid chromosome numbers. In: Jain SM, Sopory SK, Veilleus RE (eds)in vitro haploid production in higher plants, vol 1. Kluwer, Dordrecht, pp Reinert J, Bajaj YPS (1977) Anther culture: haploid production and its significance. In: Reinert J, Bajaj YPS (eds) Applied and fundamental aspects of plant cell, tissue and organ culture. Springer, Berlin Heidelberg New York pp SAS Institute (1988) SAS/STAT user s guide. Release SAS, Cary, NC, pp Van Geyt JPC, D Halluin K, Jacobs M (1985) Induction of nuclear and cell divisions in microspores of sugarbeet (Beta vulgaris L.). Z Pflanzenzücht 95: Van Geyt JPC, Speckmann GJ, D Halluin K, Jacobs M (1987)In vitro induction of haploid plants from unpollinated ovules and ovaries of the sugar beet (Beta vulgaris L.). Theor Appl Genet 73: Wan Y, Duncan DR, Rayburn AL, Petolino JF, Widholm JM (1991) The use of antimicrotubule herbicides for the production of doubled haploid plants from anther derived maize callus. Theor Appl Genet 81: Zhao J, Simmonds DH (1995) Application of trifluralin to embryogenic microspore cultures to generate doubled haploid plants inbrassica napus. Physiol Plant 95: Zhao J, Simmonds DH, Newcomb W (1996) High frequency production of doubled haploid plants ofbrassica napus cv. Topas derived from colchicine-induced microspore embryogenesis without heat shock. Plant Cell Rep 15: